HT-2020-000448 - [2024] EWHC 1185 (TCC)

DBS claims the cost of £2,050,916, apportioned on a 50/50 split between R0 and R1. The cost relates to a ‘Tiger Team’ provided by CGI.

In Closing, DBS contended that this cost related to the task of ‘carefully determining the current state of the software across the estate and identifying the best strategy for updating across the Solution as a whole’. In the Updated Schedule of Loss, DBS referred to this as ‘a preliminary analysis of the work required to update the software’. In other words, the costs claimed are not actually the costs of upgrading the particular items of software which were, in accordance with the spreadsheet, not at N-1 level.

The costs of the preliminary analysis are identified by Mr Hain as comprised within a Purchase Order dated 30 June 2021, which shows that the work paid for related an initial period of 6 weeks and then a further 8.5 months (total 10 months) of the ‘Tiger Team’. The date of Purchase Order coincides with the preceding Proposal presented by CGI dated 3 June 2021 (which contains tables of costs relating to the initial 6 week and overall 10 month periods referred to in the Purchase Order).

The ‘preliminary analysis’ which forms the basis of the claim was identified by DBS within its response to TCS’s RFI dated 16 December 202 as a particular spreadsheet, dated 11 October 2021. TCS takes the not unreasonable point that, on its face, the generation of the spreadsheet said to be the preliminary analysis which is the subject matter of the claim was produced at a point about 36% of the way into the overall 10 month period claimed on the face of the PO. On any view, the claim for remainder (74%) of the 10 month period is entirely unexplained by reference to the claimed ‘preliminary analysis’.

The further, significant, difficulty with DBS’s claim is that it is obvious from a review of the Proposal that the scope of the review undertaken by CGI was significantly broader than dealing simply with upgrading software which was, as at September 2020, not at N-1, i.e. the specific breach established. This can be seen, for example, in the following graphic:

The breach of which complaint is made falls within just the first of the 5 ‘challenges’ which the Tiger Team were, according to the Proposal, tackling. This is also clear from the ‘Key Question’/’Further Steps and ‘Activities’ tables:

None of these questions were necessitated by, or even related to, the breach relating to N-1.

It is also clear from the ‘preliminary analysis’ spreadsheet itself that, even in relation to the more limited question of upgrading the software, DBS was concerned with the broader question of what upgrades were required to the software in order to achieve ‘cloud readiness’. This is not the same as considering what is required to upgrade to N-1. So, the summary table for R1 within the preliminary analysis showed:

That the focus of just part of the overall workscope was related to broader issue of cloud readiness, rather than the fact of N-1 itself, was also reflected in CGI’s paper dated March 2022 entitled ‘R1 Data Integration (Informatica)’ which stated:

I consider that the costs claimed by reference to the 10 month consultancy project undertaken by CGI bear no direct relation to the costs that would have been incurred by reference to the actual breach which has been established. DBS incurred these costs as part of its wider business strategy and would have done so whether or not the software was maintained at N-1. Whilst I can see that conceptually, some part of the cost of this project may have been increased by the N-1 sustainment issue, no attempt whatsoever has been made by DBS to carve out from these costs those which relate to that part of the workstream which might properly be linked (or equivalent) to more limited analysis the breach would have given rise to, or the truly additional costs caused by the complaint. There is, for example, no evidence from anyone from CGI who might have been able to provide some evidence enabling an assessment along these lines.

Even if this were not of itself fatal to DBS’s claim, I also note that the preliminary analysis itself demonstrates that which might be common sense, namely that (albeit through the lens of cloud readiness), significantly more of the legacy R0 software required removal or upgrade than R1 (i.e. 41 v 71) :

DBS have simply split the costs claimed between R1 and R0 on the basis of Ms Graves’ evidence that this split was an assumption considered to be ‘fair’ which came from the finance and business teams. This was not checked with CGI (Day 15/35). I have already found that such part of the costs which related to R0 are not recoverable following the wrongful removal of R1-D. There is no basis to think that simple 50/50 split is right in circumstances where the evidence seems to show (albeit against DBS/CGI’s different criteria relating to cloud readiness) significantly more for R0 software required upgrade than R1.

In the circumstances, DBS has failed to establish that the costs claimed have any or any sufficient causal connection with the breach established in respect of R-1 Sustainment.

The Counterclaim at item 15 of the Updated Schedule of Loss fails.